The Biology of Invasive Alien in Canada. 3. (Moq.) Sauer var. rudis (Sauer) Costea & Tardif

Mihai Costea1, Susan E. Weaver2, and François J. Tardif3

1,3Department of Agriculture, Crop Science Building, University of Guelph, Guelph, Ontario, Canada N1G 2W1 (e-mail: [email protected] and [email protected]); 2Reseach Station, Agriculture and Agri-Food Canada, Harrow, Ontario, Canada NOR 1G0 (e-mail: [email protected]). Received 7 June 2004, accepted 21 October 2004.

Costea, M., Weaver, S. E. and Tardif, F. J. 2005. The Biology of Invasive Alien Plants in Canada. 3. Amaranthus tubercula- tus (Moq.) Sauer var. rudis (Sauer) Costea & Tardif. Can. J. Plant Sci. 85: 507–522. This annual dioecious weed was found in 2002 and 2003 infesting fields in southwestern Ontario, and it was collected in 1992 from waste places in British Columbia. It is a major weed problem in field crops in the mid-western , where it has become increasingly difficult to control during the past 10 yr. Morphological differences between Amaranthus tuberculatus var. rudis and var. tuberculatus are presented. A review of the biological information published is provided. Plants exhibit high phenotypic plasticity and genetic vari- ability. Emergence is prolonged, growth rapid, and female plants produce a large number of viable seeds that contribute to a per- sistent seed bank. Amaranthus tuberculatus var. rudis has developed multiple resistance to triazine and acetolactate synthase- and protoporphyrinogen-inhibiting . Airborne pollen can travel significant distances and A. tuberculatus var. rudis may hybridize with other noxious Amaranthus spp. transferring resistance or other traits.

Key words: Amaranthus tuberculatus var. rudis, AMATA, Amaranthus rudis, common waterhemp, weed biology, invasive alien

Costea, M., Weaver, S. E. et Tardif, F. 2005. Biologie des espèces exotiques envahissantes au Canada. Amaranthus tubercu- latus (Moq.) Sauer var. rudis (Sauer) Costea & Tardif. Can. J. Plant Sci. 85: 507–522. Cette adventice annuelle dioïque a été découverte en 2002 et en 2003 dans des champs de soja du sud-ouest de l’Ontario. On en avait déjà prélevé des spécimens en 1992, dans des endroits désolés de la Colombie-Britannique. Cette plante pose un grave problème dans les cultures du centre-ouest des États-Unis où elle est devenue de plus en plus difficile à contrôler au cours des dix dernières années. Suivent les différences mor- phologiques entre Amaranthus tuberculatus var. rudis et var. tuberculatus. L’article passe aussi en revue les données biologiques. La plante montre un phénotype d’une grande plasticité et un génotype très variable. Sa levée est prolongée, sa croissance rapide et les plants femelles produisent un grand nombre de semences viables qui concourent à créer un réservoir de graines tenace. Amaranthus tuberculatus var. rudis a acquis une résistance multiple à la triazine et à l’acétolactate synthase ainsi qu’aux herbi- cides inhibant le protoporphyrinogène. Le pollen véhiculé dans l’air parcourt de grandes distances et A. tuberculatus var. rudis s’hybride avec d’autres espèces nuisibles du genre Amaranthus pour leur transférer la résistance aux herbicides et d’autres carac- tères.

Mots clés: Amaranthus tuberculatus var. rudis, AMATA, Amaranthus rudis, amarante rugueuse, biologie des mauvaises herbes, plante exotique envahissante

1. Name and Generic Status tuberculatus belongs to subgenus Acnida, which comprises Amaranthus tuberculatus (Moq.) Sauer var. rudis (Sauer) 10 dioecious species (Mosyakin and Robertson 1996). For Costea & Tardif (Costea and Tardif 2003b) — Synonym: the etymology of the generic name see Costea and Tardif Amaranthus rudis Sauer—common water-hemp (2003a). “Tuberculatus” refers to the fruit surface, which (Darbyshire 2003), common waterhemp [Weed Science when dry is wrinkled and may appear as covered with tuber- Society of America (WSSA) 2004]; amarante rugueuse cles. “Rudis” alludes to the tough nature of the plants. (Darbyshire 2003). Bayer code: AMATA. Amaranthus The species has had nomenclatural and taxonomic com- tuberculatus (Moq.) Sauer var. tuberculatus—tall water- plications (see Pratt and Clark 2001). Riddel (1835) was the hemp (Darbyshire 2003), tall waterhemp (WSSA 2004); first to name A. tuberculatus as A. altissimus and A. acnide tuberculée (Darbyshire 2003). Bayer code: miamiensis. Regrettably, by providing two names, and by AMATU. , family, Amarantacées. stating that these names were “temporary”, Riddell (1835) The genus Amaranthus consists of about 70 species that invalidated them (see Greuter et al. 2001). Later, Moquin- are distributed worldwide. The approximately 40 species Tandon (1849) named the species Acnida tuberculata, which that occur in are mostly native, weeds or cul- Sauer (1955) recircumscribed to Amaranthus, together with tivated ornamentals, pseudocereals or vegetables, and only a the other 10 dioecious species. Amaranthus rudis was initial- few are introduced (Costea et al. 2001a, b). Amaranthus ly called Acnida tamariscina Nuttall. Since the type speci- 507 508 CANADIAN JOURNAL OF PLANT SCIENCE men proved to be a hybrid between A. rudis and some unde- leafy or leafless. of male flowers are 1.5–2 mm long, termined amaranth, Sauer (1955) named plants with dehis- with midrib extended into an acuminate apex, those of female cent fruits and one developed tepal Amaranthus rudis. Sauer flowers are 1.5–2.5 mm long, with a more prominent midrib. (1955) recognized two water-hemp species, A. tuberculatus Male flowers have five unequal tepals (i.e., sepals and petals found from Indiana east to Ohio and A. rudis found from are not differentiated); the outer 3 tepals are ca. 3 mm long, south to Texas, the species being sympatric in acuminate, with conspicuous, excurrent midveins; the inner , Illinois and Iowa. Pratt and Clark (2001) included tepals, ca. 2.5 mm long, obtuse or emarginate. Female flow- A. rudis in synonymy of A. tuberculatus on the basis that ers with 1 or 2 tepals, one rudimentary and one ca. 2 mm long, they represent the morphological extremes of a single narrow-lanceolate, acuminate. Fruit about 1.5 mm long, trans- species continuum. Costea and Tardif (2003b) supported versal (circumscissile) dehiscent at the middle, rugose, often their treatment as a single species, but proposed their recog- reddish. Seeds ca. 1 mm long, elliptic to obovate, with a nition at the varietal level. Historically, Uline and Bray prominent, conical hilum, dark reddish-brown. (1895) were the first to observe that “there appears to be but Grant (1959) reported a chromosome count of 2n = 32 for one polymorphous species”, but they accepted the two A. tuberculatus var. rudis from Iowa. From the same popu- species as varieties under Acnida tamariscina (Moq.) Wood. lation, a female plant was triploid (2n = 48). The dioecious Taking into account the fact that seedlings in many condition is apparently not associated with heteromorphic Amaranthus species are more or less similar, the differences chromosomes (Grant 1959). observed between the seedlings of A. tuberculatus var. rudis and var. tuberculatus are quite significant (see section 2b). (b) Distinguishing Features—Amaranthus tuberculatus is These differences, with the other evidence from morphology the only dioecious Amaranthus species in Canada and it can (see section 2a) and ecology (see section 5c), support their be easily distinguished at maturity from the monoecious recognition at varietal rank. In sympatric areas, diagnostic species by the presence of only unisexual flowers on an traits may segregate as a result of hybridization forming a individual plant. During the vegetative stage, it can be rec- unique and inseparable morphological, biological and genet- ognized by the usually narrower compared to A. ic complex (Pratt and Clark 2001; Costea and Tardif 2003b). retroflexus L., A. powellii S. Watson and A. hybridus L. An All previously published information pertaining to identification key of Amaranthus spp. from Canada can be Amaranthus rudis will be summarized below as var. rudis. found in Costea and Tardif (2003b). Following Pratt and Clark (2001), some recent studies have Amaranthus tuberculatus var. rudis can be distinguished failed to make the distinction between A. tuberculatus var. from var. tuberculatus as follows: rudis (common water-hemp) and A. tuberculatus var. tuber- Seedlings with the hypocotyl 2.5–5 mm long; cotyledons culatus (tall water-hemp). Authors have instead referred to variable in shape and size, 12–14 × 2–4 mm, ovate to linear- their plant material as “waterhemp”, A. tuberculatus. These lanceolate; first leaves ovate-lanceolate, oblong or oblong- studies have nevertheless been included in this review since it elongate 15–20 × 5–8 mm (Fig. 1 E1). Female flowers with is very likely that they referred to var. rudis, which is a more 1 or 2 tepals, lanceolate or linear (Figs. 1A, 2D). Fruit is troublesome weed than var. tuberculatus. For the sake of clar- dehiscent. (Figs. 1B, 2A) ...... A. tuberculatus var. rudis ity, we therefore recommend, if possible, that authors specify Seedlings with the hypocotyl 0.5–1.5 mm long; cotyle- in the future which A. tuberculatus variety they studied. dons uniform, 6–8 × 2.5–3 mm, ovate-elliptic; first leaves ovate to elliptic 12–18 × 9–12 mm (Fig. 1E2). 2. Description and Account of Variation Female flowers without tepals. Fruit is indehiscent. (a) Description—The following description is based primar- ...... A. tuberculatus var. tuberculatus ily on observations of populations and analysis of herbarium collections from Canada and the United States made by the (c) Intraspecific Variation— The dioecious condition authors and is supplemented with information taken from allows an even higher level of polymorphism compared to the taxonomic literature (Sauer 1955; Pratt and Clark 2001; monoecious (see Costea et al. 2004). Costea and Tardif 2003b). Amaranthus tuberculatus var. rudis exhibits extensive Plants are annual, with a taproot, and reproduce only by morphological variation at both the population and individual seeds. The hypocotyl of seedlings is 2.5–5 cm long, glabrous. levels, including variation in the overall size of plants, the pat- Cotyledons are variably shaped, ovate to linear-lanceolate, tern of stem branching and its colour (from green to red), the 12–14 long and 2–4 mm wide. The first leaves are ovate- shape of leaves, and the shape and colour of . lanceolate, oblong to oblong-elongate, with 2–4 pairs of sec- ondary veins. Stems of mature plants are erect, (5–) 20–200 (d) Illustrations—Colour photographs of entire plants are (–300) cm in height, glabrous or with sparse hairs. Leaves are available online at http://www.wssa.net and http://www. long petioled, ovate, rhombic-oblong to lanceolate-oblong, weedscience.org. The plant, seedlings, details of flowers, 2–10 long and 1–3 cm wide; the upper leaves are reduced and fruits and seeds are illustrated in Figs. 1, 2 and 3. narrow. Male and female flowers occur on separate plants (dioecious). They are grouped in axillary cymes, which are 3. Economic Importance and Environmental Impact further arranged in terminal indeterminate and dense inflores- (a) Detrimental—There are relatively few crop loss studies cences (thyrses). Terminal inflorescences are usually 10–20 available, but these findings suggest that yield losses in soy- cm long, unbranched, or with numerous panicled branches, bean [Glycine max (L.) Merr.] and corn (Zea mays L.) from COSTEA ET AL. — AMARANTHUS TUBERCULATUS 509

Fig. 1. Amaranthus tuberculatus var. rudis. A. Lower part of plant, B. of female plant. Both scale bars = 4 cm.

A. tuberculatus var. rudis interference are similar to or losses from similar densities of A. tuberculatus var. rudis greater than those produced by other weedy Amaranthus varied from 27 to 63%, depending on year and location spp. (reviewed by Costea et al. 2004). In Illinois, dense pop- (Bensch et al. 2003). Despite its later time of emergence, A. ulations of A. tuberculatus var. rudis (89 to 360 plants m–2) tuberculatus var. rudis caused greater yield losses than A. that emerged at the soybean unifoliate stage reduced soy- retroflexus, but was not as competitive as A. palmeri bean yield by an average of 43% if they remained all season, (Bensch et al. 2003). Comparable results were reported from and by 13% if they were removed 4 wk after emergence Iowa, Illinois and Missouri (Battles et al. 1998; Jones et al. (Hager et al. 2002b). In Kansas, estimated soybean yield 1998; Pfeifer et al. 2001, respectively). 510 CANADIAN JOURNAL OF PLANT SCIENCE

Fig. 2. A–D. Morphology of fruit and flowers of Amaranthus tuberculatus var. rudis. A. Female flower, B. Fruit, C. Tepals (t) and bracts (br) of female flowers, D. Tepals (t) and bracts (br) of male flowers, E. Seedlings of: 1. A. tuberculatus var. rudis, 2. A. tuberculatus var. tuberculatus. Longer scale bar (= 1 mm) for A, B, C, D; shorter scale bar (=1 mm) for E only.

In corn, season-long interference from dense populations conditions (Costea and Tardif 2003c; Costea et al. 2004). of A. tuberculatus var. rudis (60 to 300 plants m–2) reduced The same may be true about the allelopathic effects that yield in Illinois by 74% in 2 yr of the study, and 11% in the have been documented for other Amaranthus spp. [reviewed third year (Steckel and Sprague 2004). Amaranthus tuber- by Costea and Tardif (2003c), Costea et al. (2004)]. culatus var. rudis removed before corn had reached the 6- growth stage, or emerging after the corn 10-leaf stage, (b) Beneficial—Beneficial uses of this plant have not been caused no yield loss. Steckel and Sprague (2004) found that explored. Potentially, some of the uses of other Amaranthus A. tuberculatus var. rudis interference increased with mois- spp., as forage, vegetables, phytoremediation of contaminat- ture stress, and Cordes and Johnson (2003) suggested that A. ed sites, etc. (Costea and Tardif 2003c; Costea et al. 2004), tuberculatus var. rudis competes strongly with corn for may apply to A. tuberculatus var. rudis as well. nitrogen and soil moisture. Pollen grains are strongly allergenic (Lewis et al. 1983). (c) Legislation—There is no official designation in Canada Although there are no reports of cattle poisoning following or the United States (Invaders Database System 2003), and ingestion, this is a possibility since Amaranthus spp. are this issue should probably be addressed by the competent known to accumulate nitrates and oxalates under certain institutions. COSTEA ET AL. — AMARANTHUS TUBERCULATUS 511

Fig. 3. Amaranthus tuberculatus var. rudis. A. Fruit (arrows indicate dehiscence line, scale bar = 0.5 mm), B. Seed (scale bar = 0.5 mm), C. Pollen grain (scale bar = 10 µm), D. Flower (scale bar = 0.5 mm). h = hilum, t = tepal, b =

4. Geographical Distribution western Ontario in 2002 and 2003. In addition, a few speci- The geographical distribution in Canada (Fig. 4) is based on mens were collected in 1992 from waste places in Burnaby, field surveys and herbarium material. Amaranthus tubercula- British Columbia (Costea and Tardif 2003b), where apparent- tus is native to North America (Sauer 1955). Although A. ly it has not persisted, since it was not mentioned by Douglas tuberculatus var. tuberculatus is most likely native to south- et al. (1998). Amaranthus tuberculatus var. tuberculatus ern Ontario and western Québec, var. rudis has probably been occurs in Canada only on riverbanks, on the margins of lakes introduced in Canada from the United States. This is support- and ponds; it has not been collected from ruderal or agricul- ed by the fact that A. tuberculatus var. rudis has so far been tural fields. found only in agricultural fields from Petrolia (Lambton Co.), In the United States, A. tuberculatus var. rudis occurs in Cottam (Essex Co.) and Walkerton (Huron Co.) in south- 35 states, being notably absent from the western region 512 CANADIAN JOURNAL OF PLANT SCIENCE

Fig. 4. Distribution of Amaranthus tuberculatus in Canada. [ACAD, ALTA, BH, BRS, DAO, HAM, MMMN, MT, MTMG, NSPM, OAC, OTT, QFA, QK, QUE, SASK, SFS, TRTE, TUP, UAC, UBC, USAS, UWO, UWPG, V, WAT, WIN and WIS; herbarium abbreviations according to Holmgren et al. (1990)] A. General distribution in Canada, B. Distribution in southern Ontario. Dots represent A. tuberculatus var. tuberculatus, and star-shapes A. tuberculatus var. rudis.

(Sauer 1955; Sauer 1972; USDA NRCS 2004), where it is 5. Habitat likely to spread in the future. It has also been introduced to (a) Climatic Requirements—Amaranthus tuberculatus var. Europe (Aellen 1959; Joel and Liston 1986; Costea 1996). rudis occurs over a wide climatic range. The ecological COSTEA ET AL. — AMARANTHUS TUBERCULATUS 513 preferences of this taxon for temperature, water, light and Plants have an indeterminate growth habit. Stem morphology nitrogen content of the soil can be summarized as follows: (length, diameter, orientation and branching pattern) is thermophyte, hygrophyte to mesophyte, heliophyte and extremely variable depending on environmental conditions nitrophilous. and the response of plants to injury. Under unfavorable con- ditions, the shoot may not exceed 0.1 m, whereas in favorable (b) Substratum—Amaranthus tuberculatus var. rudis can circumstances, it may reach 3 m in length. Mechanical fac- tolerate a broad range of soil types and textures, but it tors, such as trampling or cutting, determine formation of sec- prefers those that are well-drained and rich in nutrients, e.g., ondary axes, and in the former case, plants may adopt a nitrophilous alluvial sands or substrates that are suitable for prostrate habit. Plants can regenerate from axillary buds if many field vegetables or row crops. The range of acceptable they are cut above the cotyledons. Cole and Holch (1941) soil pH varies from 4.5 to 8. The plant can easily tolerate the reported that roots of A. tuberculatus var. tuberculatus grow- anaerobiosis produced by temporary flooding (USDA, ing in a fine silt loam soil in Nebraska, reached about 70 cm NRCS 2004). It has no salinity tolerance, and CaCO3 toler- in depth and spread laterally for more than 2 m. ance is said to be moderate (USDA, NRCS 2004). The secondary structure of roots and the base of stems of Amaranthus tuberculatus var. rudis, growing naturally on A. tuberculatus. var. rudis is characterized by the develop- conserved prairies in South Dakota, occurred only on dry ment of successive, concentric and centrifugally developing potholes with a substrate that contained on average: 14.6 kg cambial zones (Costea and DeMason 2001). These generate ha–1 nitrate, 139 kg ha–1 phosphorus and 1041.7 kg a complex structure consisting of hundreds of collateral ha–1potassium (Umbanhowar 1992). bundles as evidenced in ground tissue. The leaves and bracts have C4 anatomy (Costea 1998; reviewed by Costea and (c) Communities in which the Species Occurs—Similar to A. Tardif 2003c). The average density of stomata in mature tuberculatus var. tuberculatus, var. rudis occurs on the mar- leaves is 120 stomata mm–2 in the lower epidermis and 95 gins of fresh waters, rivers, lakes, ponds, marshes, and bogs. stomata mm–2 in the upper epidermis (Costea 1998). The However, unlike the former, var. rudis also grows in dis- morphology and structure of the small fruits (Costea et al. turbed habitats, such as roadsides, railroads, cultivated 2001c) and seeds facilitate dispersal by water, animals or fields and gardens. In Canada, it has been encountered only birds, and to a lesser extent by wind (see section 8b). in the latter type of habitats. In a soybean field near Petrolia, Embryology is similar to other Amaranthus spp. (see Costea in southwestern Ontario, its density reached 100–220 plants and Tardif 2003c; Costea et al. 2004). m–2. Other weed species with much lower densities observed at the site were: Chenopodium album L., A. retroflexus, A. (b) Perennation—Amaranthus tuberculatus var. rudis powellii and Polygonum persicaria L. In Iowa, A. tubercula- plants are summer annuals and overwinter as seeds on or tus var. rudis reached a density of 346 plants m–2 in corn and below the surface of the soil. , when no herbicides were applied and represented (c) Physiological Data—Amaranthus spp. have the C4 path- up to 51% of the total weed population consisting of way of photosynthesis (see Costea and Tardif 2003c; Costea Ambrosia artemisiifolia L., Asclepias syriaca L., et al. 2004), and exhibit the characteristic Kranz anatomy in Chenopodium album, Conyza canadensis (L.) Cronq., leaves, cotyledons (see Costea et al. 2004) and bracts (Costea Cyperus esculentus L., Hibiscus trionum L., Setaria faberi and Tardif 2003d). Although there are few physiological Herrm., S. glauca (L.) Beauv., Sida spinosa L., Solanum studies on A. tuberculatus, it may be assumed that var. rudis carolinense L., S. ptycanthum Dun., and Polygonum pensyl- has traits similar to other weedy Amaranthus spp.: high pho- vanicum L. (Felix and Owen 1999). tosynthetic rates at high temperatures and light intensity, reduced photorespiration, low CO2 compensation point, 6. History greater water use efficiency, and increased nitrogen use effi- Herbarium vouchers attest the presence of A. tuberculatus ciency (see Costea and Tardif 2003c; Costea et al. 2004). var. tuberculatus in Ontario and Québec from the end of the In a greenhouse study by Guo and Al-Khatib (2003), A. 19th and the beginning of the 20th century. For example, in tuberculatus var. rudis biomass, height and root volume Ontario: London 1880 (DAO, ROM), Thames River 1887; measured at 4 wk after emergence were higher at 25/20°C Komoka 1890 (DAO), Ottawa 1892 (DAO); in Québec: St. and 35/30°C than at 15/10°C. Plants died 9 d after being Amélie 1890 (MT), Saint Laurent 1909 (MT). In contrast, exposed to 45/40°C. Rubisco activity was greater at except for a few collections from Burnaby, British 15/10°C than at 25/10 or 35/10°C, whereas the chlorophyll Columbia, there are no herbarium records of A. tuberculatus content and fluorescence varied inversely. var. rudis in Canada. Its recent occurrence in agricultural Horak and Loughin (2000) conducted a 2-yr field study in fields from southern Ontario suggests that this weed has Kansas comparing the growth rates of four Amaranthus spp. been recently introduced from the United States. in the absence of competition. Growth parameters recorded for A. tuberculatus var. rudis were: rate of height increase of 7. Growth and Development 0.11 to 0.16 cm per growing degree day; maximum relative (a) Morphology—Amaranthus tuberculatus var. rudis has the growth rate of 0.31 g g–1 d–1; net assimilation rate of 3.9 mg attributes of an invasive successful weed: ecological plastici- cm–2 d–1; specific leaf area of 160 to 205 cm2 g–1; plant vol- ty, rapid growth, production of numerous seeds over a pro- ume of 1 127 162 to 3 946 463 cm3; and dry weight of 222 longed period of time, and seed dormancy (see section 8c). to 742 g plant–1. Based on these parameters, A. tuberculatus 514 CANADIAN JOURNAL OF PLANT SCIENCE var. rudis ranked second after A. palmeri and ahead of A. female seeds are produced in dioecious species, Lemen retroflexus and A. albus L. In a similar experiment under- (1980) found that the proportion of male plants in a popula- taken in Missouri, A. tuberculatus var. rudis ranked after A. tion of A. tuberculatus var. rudis from Illinois was 0.33, palmeri, A. retroflexus and A. hybridus, based on dry weight which significantly deviates from the normal 1:1 sex ratio. measured at the end of the season (Sellers et al. 2003). The ratio of males to females in Ontario field populations Similar to other Amaranthus spp., A. tuberculatus var. in 2003 and when grown from seed in the greenhouse was rudis plants exhibit a facultative short-day flowering very close to 1:1 (Weaver, unpublished data). Menalled response. Under short-day conditions, flowering is initiated et al. (2004) reported a marginally higher male to female very early, at the cost of reduced vegetative growth and ratio in plants grown from seed receiving compost-amend- lower seed production. Under long-day conditions, plants ed treatments compared with compost-free treatments grow larger, flower later, and produce more seeds if suffi- (P = 0.061). cient time is available (see section 7d). Steckel et al. (2003) Flowers are small, wind-pollinated and plants are neces- in Illinois reported that growth of female plants of var. rudis sarily outcrossing. Male flowers have five stamens; anthers depended on time of emergence and light availability. Plants open longitudinally, and their volume is about 0.157 mm3 grown under full sun accumulated 720 g plant–1 when they (Lemen 1980). The pollen grain is spherical, 18–25 µm in emerged in late May and 350 g plant–1 when they emerged diameter, with 90–110 sunken apertures uniformly distrib- in late June. Under 40 and 68% shading, plants produced uted on its surface (Fig. 2C) (Costea 1998; Franssen et al. 550 and 370 g plant–1, respectively, when emerged in May, 2001b). The number of apertures is twice that found in and 220 and 170 g plant–1, respectively, when emerged in monoecious species (Costea 1998; Franssen et al. 2001b). June. Furthermore, under shaded conditions, more biomass These are extreme aerodynamic adaptations for wind polli- was partitioned to the leaves at the cost of reproductive nation. The numerous, uniformly distributed apertures gen- structures, resulting in higher leaf area ratios (up to 1.5-fold, erate a layer of turbulent air that decreases the friction 2.2-fold and 2.8-fold increase under 40, 68 and 99% shade between the pollen grain and the air, maximizing the dis- levels, respectively). tance pollen grains can be wind-dispersed (Franssen et al. 2001b). The pollen grain is covered with granules that (d) Phenology—Amaranthus tuberculatus var. rudis is a ensure the adherence to stigma hairs. summer annual that emerges in southern Ontario from the The perianth of female flowers is reduced to one or two beginning of June to August. In Iowa, emergence began in short tepals, and therefore the gynoecium is more exposed late May and continued through early August (Hartzler et al. than in other dioecious (e.g., A. palmeri) or some monoe- 1999). In Missouri, A. tuberculatus var. rudis emerged 7 to cious species (e.g., A. retroflexus), which have five tepals 10 d later than A. hybridus and A. albus, and 10–12 d later that are longer than the gynoecium. The gynoecium does not than A. retroflexus (Sellers et al. 2003). Flower initiation have a style and consists of two or three united carpels depends on photoperiod. Plants grown under short-day con- (Costea et al. 2001c). The two or three stigma are signifi- ditions (8 h) require 14–16 d to initiate flowering, whereas cantly longer than in monoecious species (stigmas of plants grown under long-day conditions (16 h) need approx- 3.2–4.3 mm in A. tuberculatus var. rudis, versus stigmas of imately 45 d. These data are in agreement with those for 0.9–2.1 mm in A. hybridus, A. powellii and A. retroflexus; other Amaranthus spp. (reviewed by Costea et al. 2004). Costea 1998). The stigmas are covered with 2–4 rows of Flowering and seed set continue until the first frost, which receptive hairs (Costea et al. 2001c). Stigmas of unfertilized in southern Ontario coincides with the end of October or the female flowers can persist indefinitely until pollen reaches beginning of November. them, consistent with observations on another dioecious species, A. cannabinus (Quinn et al. 2000). After fertiliza- (e) Mycorrhiza—Although Amaranthaceae have traditional- tion, the stigmas dry out. ly been considered non-mycorrhizal, vesicular-arbuscular In two populations from Illinois, female plants had at mycorrhizae have been observed under certain conditions in least one male plant within 1 to 6 m distance (Lemen 1980). several other Amaranthus spp. (reviewed by Costea et al. However, the maximum distance that pollen can travel is 2004). much further. We determined the downwind distance of pollen transfer with the following equation (Kyaw Tha Paw 8. Reproduction and Hotton 1989): (a) Floral Biology—The inflorescence of Amaranthus spp. is fairly complex (reviewed by Costea et al. 2004). Flowers L = UH/Vset (1) are unisexual (diclinous), and in the dioecious species, male and female flowers occur on separate plants. Steinau et al. where L is the downwind distance of pollen transfer (m); U (2003) observed that a hybrid between A. tuberculatus var. is mean wind speed (m s–1); H is the height at which pollen –1 rudis and A. hybridus had predominantly female flowers, is released (m); and Vset is the settling velocity (m s ). After but also a small male inflorescence making such plants their release, pollen grains fall towards the ground with a monoecious. The flowers of both male and female plants certain settling velocity (Vset), which is determined by the develop in numerous dense cymes, which are further aerodynamics of the pollen grain and gravity. Settling veloc- arranged in spikes (spiciform branches). Although Murray’s ities can be determined using the equation of Stokes (1940) results indicated that an equal number of male and (Gregory 1973): COSTEA ET AL. — AMARANTHUS TUBERCULATUS 515

ρ ρ 2 µ –1 Vset = 2/9 ( p – a) d / (2) corn 6-leaf stage produced 950 seeds plant , and plants emerging after the corn 8-leaf stage failed to reproduce ρ –1 ρ where p is pollen density (g cm ); a is atmospheric densi- (Nordby and Hartzler 2004). Seed production by plants ty; d is pollen diameter; and µ is dynamic viscosity (approxi- emerging in late May in Illinois declined from over 1 million mately 1.8 × 10–4 g cm–1 s–1). Assuming that pollen grains are seeds per plant in full sunlight, to 401 200 seeds plant–1under spherical, without any ornamentation, and that pollen density 68% shade and 8 seeds under 99% shade (Steckel et al. 2003). is 1 g cm–1 (Kyaw Tha Paw and Hotton 1989), the calculated Seeds from North Dakota and Illinois weigh 0.19–0.27 g per value of Vset for A. tuberculatus var. rudis would be between 1000 seeds (Stevens 1932; Lemen 1980; Sellers et al. 2003). 0.0185 and 0.021 m s–1, depending on pollen size. The size and weight of the seeds are likely to vary among dif- Based on Eq. 1, if male plants were 0.5 m tall, and the ferent populations, among individuals within populations and wind velocity 10 km h–1, pollen grains could reach a dis- even from the same plant (see Costea et al. 2004). tance of 35–50 m. If plants were 1 m tall and the wind veloc- Seed dispersal is accomplished by water, farm machinery, ity 40 km h–1, pollen grains could travel 300–325 m. These spreading of manure and compost, birds, animals, and to a estimates do not take into account the effect of the plant lesser extent by wind, as with other Amaranthus spp. canopy, which is likely to significantly reduce the distance (Costea et al. 2004). Dispersal by water (streamlets pro- of pollen transfer, or the aerodynamic ornamentation of duced on the soil by rain, surface irrigation, and rivers) may pollen grains, which is likely to increase the distance of be significant since both fruits and seeds float, and plants pollen transfer. These simple estimations show that pollen prefer the proximity of water. There is no information can be carried significant distances. However, isolated regarding the survival capacity of seeds in the digestive tract female plants will not form seeds, and this may be viewed of animals or during composting. Nevertheless, judging by as a significant cost of dioecism. the similar structure of the seed coat, it can be speculated Inbreeding is often considered a specialization of annual that these values are comparable to those observed in other weeds and pioneer species (e.g., Baker and Stebbins 1965). Amaranthus spp. (see Costea et al 2004). For example, seeds In the case of Amaranthus spp., some dioecious species, like of A. retroflexus survived digestion by 10-mo-old lambs, A. tuberculatus var. rudis and A. palmeri, are just as suc- rumen digestion by cattle (27–45% of the seeds survived), cessful as monoecious species, such as A. retroflexus, A. ensiling followed by rumen digestion (1–6% of the seeds powellii, and A. hybridus. The recombination potential of survived), and a small percentage of seeds (3.5%) survived dioecious amaranths appears to outweigh the costs of dioe- after 2 wk of windrow composting at temperatures of cism, especially when the large number of seeds produced 55–65°C (reviewed by Costea et al. 2004). by a single plant renders irrelevant the fact that initial colo- nization cannot start from a single seed (see section 8b). (c) Seed Banks, Seed Viability and —Seeds of Indeed, colonization by A. tuberculatus var. rudis has to A. tuberculatus var. rudis have a high (>80%) initial viabil- begin with several plants (a micropopulation), and not with ity (Leon et al. 2004; Steckel et al. 2004), and are likely to single plants as in the case of monoecious amaranths. form a persistent seed bank, similar to other Amaranthus Although this may be regarded as a disadvantage, after ini- spp. (Costea and Tardif 2003c; Costea et al. 2004). Burnside tial establishment, its spread will progress as fast as in et al. (1996) reported that 1–3% of A. tuberculatus var. rudis monoecious species. seeds germinated after 17 yr of burial at a soil depth of approximately 20 cm in Nebraska (viability of remaining (b) Seed Production and Dispersal—Unlike the monoecious seeds was not determined). Buhler and Hartzler (2001) amaranths, only female plants of A. tuberculatus var. rudis found that 11% of seeds maintained their viability after 4 yr produce seeds, and therefore seed production at the popula- of burial in the upper 5 cm of soil in Iowa. In Iowa, tion level may be lower than in the former species. However, Amaranthus tuberculatus var. rudis was one of the domi- female plants of dioecious species may have a greater capac- nant species appearing on land brought back into crop pro- ity for seed production because they do not partition resources duction after 8 yr under the conservation reserve program to male flowers as in monoecious amaranths (see Costea et al. (Felix and Owen 1999). 2004). Comparison of rates of increase of populations of dioe- The seed bank of A. tuberculatus var. rudis in agricultur- cious and monoecious amaranths deserves further attention. al fields from Iowa ranged in size from 1910 to 64 160 seeds A single female plant of A. tuberculatus var. rudis normally m–2, and comprised up to 90% of the total seed bank (Buhler produces between 35 000 and 1 200 000 seeds (Stevens 1932; et al. 2001). Comparatively, seed banks of A. tuberculatus Sellers et al. 2003; Steckel et al. 2003; Hartzler et al. 2004). var. rudis from natural prairie wetlands in Iowa were much Fecundity declines, along with biomass and height, as time of lower, averaging 181 seeds m–2 (Galatowitsch and van der emergence is delayed (Steckel et al. 2003; Hartzler et al. Valk 1996). The size of the seed bank is influenced by 2004; Nordby and Hartzler 2004). Female plants emerging numerous factors: the tillage system, various cultural prac- with soybean at four sites in central Iowa produced on aver- tices, including weed management practices, and crop rota- age 300 000 seeds plant–1, whereas those emerging 50 d after tion, which in turn affect the number of seed produced, their planting produced 3000 seeds plant–1 (Hartzler et al. 2004). In vertical distribution in the soil profile, and the probability of corn, in Iowa the earliest emerging A. tuberculatus var. rudis seed survival and germination. Buhler et al. (2001) charac- plants produced 48 400 seeds, whereas those emerging at the terized the size and depth of the seed bank of A. tubercula- 516 CANADIAN JOURNAL OF PLANT SCIENCE tus var. rudis in a field in central Iowa throughout a 5-yr Acnida, as well between var. rudis and monoecious species crop rotation consisting of legume/grass hay → corn → soy- from the subgenus Amaranthus (Thellung 1914; Wetzel et al. bean → corn → oat. The seed bank was greatest during the 1999b; Franssen et al. 2001a; Costea and Tardif 2003b; hay/oat phases, but declined rapidly during the corn and Steinau et al. 2003; Costea et al. 2004). First generation soybean phases. They attributed the decline in corn and soy- hybrids between A. tuberculatus (variety unspecified) and A. beans to moldboard plowing, which distributed the seeds hybridus were dioecious (Trucco et al. 2004). Natural over a greater soil depth, and inter-row cultivation, which hybridization rates between A. tuberculatus (variety unspeci- reduced A. tuberculatus var. rudis survival and seed return, fied) and A. hybridus were as high as 5.9% when the former and the increase in hay and oat crops, to its ability to emerge species was used as the male parent, and 0.7% when it was after harvest and produce numerous seeds that remained used as a female parent (Trucco et al. 2004). Franssen et al. concentrated near the soil surface in the absence of plowing. (2001b) found that the number of apertures on the pollen Comparable results have been reported for other grains of a hybrid between A. tuberculatus var. rudis and A. Amaranthus spp. (reviewed by Costea et al. 2004). palmeri was intermediate between the parents. Other mor- The fraction of the A. tuberculatus var. rudis seed bank phological characters may not be intermediate between the emerging in a single year in central Iowa varied from 1 to two parents (Steinau et al. 2003), as has also been reported for 22%, and was greater in years of higher rainfall and temper- hybrids between monoecious species of the section ature (Hartzler et al. 1999; Buhler and Hartzler 2001). Amaranthus (Costea and Tardif 2003b; Costea et al. 2004). Common water-hemp emergence begins later and occurs The presence of morphological characteristics absent in par- over a more prolonged period of time compared to most ent species may be explained by the formation of polymor- annual weeds (Hartzler et al. 1999). Most seedlings emerge phic DNA fragments resembling transposon-like elements from near the soil surface, probably because the hypocotyl (Wetzel et al. 1999b; Steinau et al. 2003). First generation can elongate only 0.5–3.5 (–5) cm. Seedling emergence was hybrids between dioecious and monoecious, or between 102% greater in areas marked by tractor wheel traffic than monoecious Amaranthus species, have reduced fertility, low in untracked areas (Jurik and Zhang ShuYu 1999). vigor and slow growth, distorted leaves, abnormally shaped The minimum temperature for germination was 10°C for inflorescences, and numerous, densely packed bracteoles. populations from Iowa (Leon et al. 2004), over 15/10°C for Such characters, together with the presence of parent species populations from Kansas (Guo and Al-Khatib 2003), and 15 in the field may serve as a preliminary identification method to 20°C for populations from Illinois (Steckel et al. 2004). for hybrids. Two molecular datasets, restriction enzyme poly- The optimum temperature was 33–35°C, above which ger- morphisms within the internal transcribed spacers (ITS) of the mination declined (Guo and Al-Khatib 2003; Leon et al. ribosomal DNA (Wetzel et al. 1999a) and amplified fragment 2004). Approximately 20% of seeds germinated at 45/40°C length polymorphism (AFLPs) (Wetzel et al. 1999b), have and no seeds germinated at 50/45°C (Guo and Al-Khatib not proven useful for identification of Amaranthus hybrids. In 2003). Alternating temperatures increased germination contrast, DNA content analysis has been successfully used to compared to constant temperatures, and the greater the tem- distinguish between A. hybridus and A. tuberculatus (Tranel perature variation the larger the increase in germination, et al. 2002; Jeschke et al. 2003). F1 hybrids exhibited nuclear with the optimum amplitude 18°C (Leon et al. 2004; Steckel DNA contents equal to the mean of the two parents (e.g., A. et al. 2004). hybridus 1.04 pg and A. tuberculatus 1.34 pg) (Tranel et al. Seeds of Amaranthus spp. shed on the soil surface or buried 2002; Jeschke et al. 2003). at different depths, undergo cyclical changes in dormancy, Amaranthus spp. often grow in large mixed populations regulated primarily by seasonal variation in temperature of two or three species. As a consequence of hybridization, (reviewed by Costea et al. 2004). Most freshly harvested A. introgression occurs and hybrid swarms result. When tuberculatus var. rudis seeds are dormant but will germinate hybrids between A. tuberculatus and A. hybridus were back- after wet stratification at 4°C for 12 wk (Leon and Owen crossed to A. hybridus, the progeny include both monoe- 2003; Leon et al. 2004). Seed dormancy and germination are cious and dioecious types and their DNA contents were regulated by phytochrome. Germination is stimulated by intermediate (Tranel et al. 2004). From a practical point of exposure to red light and the effect is reversible by treatment view, hybridization is one route by which resistance to her- with far-red light (Leon and Owen 2003). Temperature influ- bicides can be transferred between different Amaranthus ences light sensitivity, and high temperatures (36°C) promot- spp. Such transfers of herbicide resistance have been docu- ed germination of chilled seeds even after they were treated mented from A. palmeri to A. tuberculatus var. rudis with far-red light (Leon and Owen 2003). These results are in (Wetzel et al. 1999b; Franssen et al. 2001a), and from A. agreement with those reported for other Amaranthus spp. hybridus to A. tuberculatus var. rudis (Tranel et al. 2002). (Costea and Tardif 2003c; Costea et al. 2004). 10. Population Dynamics (d) Vegetative Reproduction—Amaranthus tuberculatus Because of its late emergence, A. tuberculatus var. rudis is var. rudis does not reproduce vegetatively. likely to produce only one generation per year. The plant rapidly invades disturbed areas because of copious seed pro- 9. Hybrids duction and the formation of persistent seed banks that build Natural hybrids have been reported between A. tuberculatus up very quickly in the absence of control. During the past var. rudis and other dioecious species from the subgenus decade, A. tuberculatus var. rudis has gradually emerged as COSTEA ET AL. — AMARANTHUS TUBERCULATUS 517 a dominant weed species in corn and soybeans throughout fen. These herbicides provided 75–90% control of A. tuber- the mid-western United States. An increase in the frequency culatus var. rudis in Illinois, with lactofen and fomesafen and severity of infestations of A. tuberculatus var. rudis has more effective than (Hager et al. 2003). Control been caused by changes in cultural practices and weed man- by all three herbicides declined with time after treatment, as agement, and by the rapid development of herbicide-resis- some plants re-grew from lower leaf nodes and new tant biotypes (see sections 11 and 12). seedlings germinated. Sweat et al. (1998) and Mayo et al. Comparative growth studies in Kansas, suggested that A. (1995) reported that A. tuberculatus var. rudis could also be tuberculatus var. rudis is competitively inferior to A. controlled postemergence by imazamox, imazethapyr, chlo- palmeri, but superior to A. retroflexus (Horak and Loughin rimuron and thifensulfuron, provided the populations were 2000). In Missouri, A. tuberculatus var. rudis ranked after A. not resistant to herbicides that inhibit the enzyme acetolac- palmeri, A. retroflexus, and A. hybridus in terms of dry tate synthase (ALS). Wide-spread resistance to ALS- weight gain and height, but had more efficient seed produc- inhibitors now makes these herbicides ineffective on many tion (by 2.0-, 1.4- and 1.4-fold greater, respectively) (Sellers A. tuberculatus var. rudis populations. et al. 2003). Triazine-resistant A. tuberculatus var. rudis In corn from Nebraska, preemergence application of S- plants in Nebraska were less competitive than triazine-sus- metolachlor plus atrazine provided season-long control ceptible plants at densities of 50, 100 and 150 plants m–2, (Nolte and Young 2002a). Other preemergence herbicides while at 300 plants m–2 they were equally competitive reported to provide control of A. tuberculatus var. rudis in (Anderson et al. 1996b). corn are pendimethalin, dimethenamid, isoxaflutole, and (Steckel et al. 2002; Peter Sikkema, University 11. Response to Herbicides and Other Chemicals of Guelph, unpublished data). Sequential applications and Amaranthus tuberculatus var. rudis is susceptible to many the addition of atrazine generally improved the level of con- selective and non-selective herbicides available for the con- trol (but see triazine resistance below). Postemergence con- trol of broad-leaved weeds in crop and non-crop areas. trol can be achieved with atrazine, primisulfuron, dicamba, However, management of A. tuberculatus (both varieties) prosulfuron plus dicamba, 2,4-D plus atrazine, diflufen- with herbicides is complicated by its prolonged emergence zopyr plus dicamba, and mesotrione (Anderson et al. 1996a; pattern, genetic variability within and between populations, Peter Sikkema, University of Guelph, unpublished data). and the ability of the species to rapidly develop resistance In sugarbeets in Minnesota, season-long A. tuberculatus after repeated exposure (Patzoldt et al. 2002). var. rudis control required three sequential applications of Amaranthus tuberculatus var. rudis emerges after most desmedipham alone or in combination with triflusulfuron- crops have been planted, so herbicides applied before plant- methyl, clopyralid and methylated seed oil, or metolachlor ing are not an effective control measure, unless they have or dimethenamid (Roehl et al. 2001). residual soil activity. Herbicides with residual activity Herbicide-resistant crops provide an opportunity to con- applied to the soil at planting will control emerging A. tuber- trol A. tuberculatus var. rudis up to 30 cm in height with the culatus var. rudis for at least part of the season. Foliar- non-selective herbicides glyphosate or glufosinate (Coetzer applied postemergence herbicides will control growing et al. 2002; Hoss et al. 2003). However, these products have weeds, but in the absence of residual activity, repeated no residual activity, and consistent, season-long control applications must be made to control new flushes. In many (>80%) of A. tuberculatus var. rudis in corn in Missouri crops, a soil-applied herbicide followed by a postemergence required two applications of glyphosate or glufosinate in herbicide is required for season-long control. Broadcast her- combination with residual herbicides (Hellwig et al. 2003). bicide treatments controlled A. tuberculatus var. rudis better Single applications were much less effective, unless com- than band treatments (Felix and Owen 1999). bined with a residual herbicide, such as atrazine in corn, or In a number of studies in the mid-western United States, sulfentrazone in soybeans (Bradley et al. 2000; Dirks et al. the most effective soil-applied herbicide for control 2000a; Johnson et al. 2000; Beyers et al. 2002; Nolte and (97–98%) of A. tuberculatus var. rudis in soybeans was Young 2002a, b). Control with single glyphosate applica- sulfentrazone (Krausz et al. 1998; Dirks et al. 2000b; Hager tions in soybeans increased as row spacing decreased from et al. 2002a; Nolte and Young 2002b; Krausz and Young 76 to 19 cm (Young et al. 2001). 2003). Other preemergence herbicides reported to control A. Throughout the mid-western United States, A. tubercula- tuberculatus var. rudis are flumioxazin, dimethenamid, S- tus var. rudis populations have developed resistance to her- metolachlor, pendimethalin, acetochlor, linuron, bicides that inhibit photosystem II (Group 5, e.g., triazines), imazethapyr, metribuzin, flufenacet plus metribuzin, and ALS-inhibitors (Group 2, e.g., sulfonylureas) and herbicides flumetsulam plus metolachlor (Sweat et al. 1998; Niekamp that inhibit protoporphyrinogen oxidase (PPO, Group 14, and Johnson 2001; Steckel et al. 2002). Preemergence her- e.g., diphenylethers) (Heap 2004). Resistance to triazine bicides should be applied as close to planting as possible, to herbicides was first reported in Nebraska in 1990 (Anderson maximize the duration of activity (Hager et al. 2002a). By et al. 1996b), and has since been discovered in Missouri, 56 d after planting, sequential herbicide applications usual- Kansas, Iowa, Illinois and Ontario (Heap 2004). Resistance ly provide better control than single treatments. to ALS-inhibiting herbicides was confirmed in 1993, and Postemergence control of A. tuberculatus var. rudis up to now occurs in Iowa, Missouri, Oklahoma, Kansas, Illinois, 10 cm in height can be achieved in soybeans by the Ohio, and Ontario (Horak and Peterson 1995; Peterson diphenylether herbicides lactofen, fomesafen and acifluor- 1999; Heap 2004). In Ontario, multiple resistance to 518 CANADIAN JOURNAL OF PLANT SCIENCE imazethapyr (Group 2) and atrazine (Group 5) was first Felix and Owen 1999). Survival, plant height, biomass, and reported from soybean fields in 2002 (Heap 2004). seed production all decline when A. tuberculatus var. rudis Resistance to the PPO (protoporphyrinogen oxidase)- emergence is delayed relative to a crop (Hartzler et al. 2004; inhibiting herbicides lactofen, acifluorfen, fomesafen and Nordby and Hartzler 2004). Nitrogen application at 120 kg sulfentrazone was observed in a biotype in Kansas in 2001, ha–1 early in the season provided a competitive advantage to after 4 yr of repeated exposure to acifluorfen (Shoup et al. corn in Nebraska, and reduced growth of A. tuberculatus 2003). This biotype also had a high level of resistance to the var. rudis and Abutilon theophrasi Medic. (Evans et al. ALS herbicides imazethapyr and thifensulfuron. Resistance 2003). Results obtained in Iowa have shown that composted to acifluorfen and lactofen was reported in two soybean swine manure applied at rate of 4000 or 8000 kg C ha–1 fields in Missouri (Li et al. 2004b). Falk et al. (2004) con- reduced the emergence of A. tuberculatus var. rudis and ducted a survey of northeastern Kansas, and found PPO other weeds, but increased their overall competitive ability resistance in half of the sampled fields, and ALS resistance (Menalled et al. 2004). in most A. tuberculatus var. rudis populations. Other cases of multiple resistance to triazines, ALS or PPO inhibitors 13. Response to Herbivory, Disease and Higher have been reported in Illinois (Heap 2004). Although true Plant Parasites resistance to glyphosate has not yet been reported in any of The response of A. tuberculatus var. rudis to various herbi- the varieties of A. tuberculatus, populations exhibit a wide vores, diseases, and higher plant parasites is unknown. range of response to this herbicide, and appear to vary in tol- However, it is very likely that A. tuberculatus var. rudis is erance (Li et al. 2004a; Patzoldt et al. 2004). affected more or less by the same nematodes, viruses, fungi Amaranthus tuberculatus (variety unspecified) populations and bacteria as other Amaranthus weed species (Costea and differ in the mechanism of resistance to the various herbicide Tardif 2003c; Costea et al. 2004). classes and in the pattern of cross-resistance to other herbi- cides within the same chemical family. In most weed species, (a) Herbivory: triazine resistance is conferred by a mutation at the site of action in the chloroplast, and is maternally inherited. Patzoldt (i) Mammals—No data. et al. (2003) surveyed A. tuberculatus (variety unspecified) populations in Illinois and found that some segregated for tri- (ii) Birds and/other vertebrates—No data. azine resistance and that the trait was nuclear encoded and not maternally inherited. These populations had a lower level of (iii) Insects—Preliminary studies on post-dispersal seed pre- resistance and less cross-resistance to triazines other than dation done at Boone, Iowa (van der Laat and Owen, per- atrazine compared to populations that were uniformly resis- sonal communication) have indicated that more or less the tant and possessed target site resistance. Patterns of cross- same insects associated with other Amaranthus spp. (Costea resistance to herbicides that inhibit the ALS enzyme also vary et al. 2004) fed on the seeds of A. tuberculatus var. rudis as from one population to another, indicating different single well. The most important post-dispersal seed predators point mutations (Lovell et al. 1996; Hinz and Owen 1997; observed were: Amara aeneopolita Casey, Anisodactylus Sprague et al. 1997a, b). Foes et al. (1998) found that a tryp- rusticus Say, Stenolophus comma (F.), Gryllus pennsylvan- tophan to leucine substitution at position 574 of the ALS gene icus Burmeister, and Harpalus pennsylvanicus De Geer. conferred broad cross-resistance to ALS-inhibiting herbi- Among these, the latter two species had the highest popula- cides, i.e., more than 10-fold resistance to sulfonylureas, imi- tions (van der Laat and Owen, personal communication). dazolinones and triazolopyrimidines. On the other hand, a substitution from serine to asparagine or threonine at position (iv) Nematodes—No data. 653 of the ALS gene resulted in resistance to imidazolinones, but not sulfonylureas or triazolopyrimidines (Tranel and (b) Diseases: Wright 2002). The physiological basis for resistance to the PPO-inhibitor herbicides acifluorfen and lactofen in Missouri (i) Fungi—Canada—No records. United States: Albugo bliti A. tuberculatus var. rudis populations involves reduced accu- (Biv.-Bern.) Kuntze: Minnesota (Preston and Dostall 1955); mulation of protoporphyrin IX (Li et al. 2004b). Oklahoma (Preston 1945); North America (Wilson 1908); Genetic variation within and between populations in Phymatotrichum omnivorum (Duggar) Hennebert: Texas response to various herbicides (Patzoldt et al. 2002), and inter- (Anonymous 1960). Additionally, var. tuberculatus was specific hybridization, and gene flow between monoecious reported to be a host of Albugo bliti: Iowa, Kansas and (Tranel et al. 2002) and dioecious Amaranthus species (Wetzel Wisconsin (Anonymous 1960); Cercospora acnidae Ellis & et al. 1999b; Franssen et al. 2001a), probably contribute to the Everh.: Wisconsin (Greene 1945; Anonymous 1960); and rapid spread of herbicide resistance in these species. Phyllosticta amaranthi Ellis & Kellerm: New York (Anonymous 1960). 12. Response to Other Human Manipulations Amaranthus tuberculatus var. rudis populations increase Biocontrol—Preliminary tests using Microsphaeropsis under no-tillage or reduced tillage cropping systems, amaranthi (Ellis & Barthol.) Heiny & Mintz (= because seeds remain near the soil surface, which promotes Aposphaeria amaranthi Ellis & Barthol.) concluded that germination and emergence (Hager et al. 1997; Hager 1998; this fungus has potential as a bioherbicide for the control of COSTEA ET AL. — AMARANTHUS TUBERCULATUS 519

A. tuberculatus. The fungal pathogen caused foliar and stem Anderson, D. D., Roeth, F. W. and Martin, A. R. 1996a. necrosis resulting in the mortality of A. tuberculatus (variety Occurrence and control of triazine-resistant common waterhemp unspecified) plants under optimal conditions (Smith and (Amaranthus rudis) in field corn (Zea mays). Weed Technol. 10: Hallett 2004a). Tank mixtures of the fungus conidia and 570–575. glyphosate had reduced efficacy (Smith and Hallett 2004b). Anderson, D. D., Higley, L. G., Martin, A. R. and Roeth, F. W. 1996b. Competition between triazine-resistant and -susceptible common waterhemp (Amaranthus rudis). Weed Sci. 44: 853–859. (ii) Bacteria—No data. Anonymous 1960. Index of plant diseases in the United States. US Dept. Agric. Handbook No. 165. Washington, DC. 531 pp. (iii) Viruses—No data. Baker, H. G. and Stebbins, G. L. (eds.) 1965. The genetics of colonizing species. Academic Press, New York, NY. 588 pp. (c) Higher Plant Parasites—No information was located. Battles, B., Hartzler, B. and Buhler, D. 1998. Effect of common waterhemp emergence date in soybeans on growth and competi- 14. Prognosis tiveness. Proc. North Centr. Weed Sci. Soc. 53: 145–146. The presence of A. tuberculatus var. rudis as a weed of arable Bensch, C. N., Horak, M. J. and Peterson, D. 2003. Interference fields represents a threat to crop production in Eastern Canada, of redroot pigweed (Amaranthus retroflexus), Palmer amaranth (A. particularly where soybeans and corn are grown. The recent palmeri), and common waterhemp (A. rudis) in soybean. Weed range expansion throughout the mid-western United States Sci. 51: 37–43. Beyers, J. T., Smeda, R. J. and Johnson, W. G. 2002. Weed and its ability to rapidly develop herbicide resistant biotypes management programs in glufosinate-resistant soybean (Glycine have caused an increase in the frequency and severity of infes- max). Weed Technol. 16: 267–273. tations. It spread from southern Illinois to become the most Bradley, P. R., Johnson, W. G., Hart, S. E., Buesinger, M. L. troublesome weed throughout the state in less than a decade and Massey, R. E. 2000. Economics of weed management in glu- (Steckel and Sprague 2004). Amaranthus tuberculatus var. fosinate-resistant corn (Zea mays L.). Weed Technol. 14: 495–501. rudis has a superior genetic recombination potential compared Buhler, D. D. and Hartzler, R. G. 2001. Emergence and persis- to monoecious amaranth species (e.g., A. retroflexus, A. pow- tence of seed of velvetleaf, common waterhemp, woolly cupgrass, ellii, etc.), and therefore a higher genetic and phenotypic vari- and giant foxtail. Weed Sci. 49: 230–235. ability. Its propensity for interspecific hybridization and Buhler, D. D., Kohler, K. A. and Thompson, R. L. 2001. Weed introgression with other Amaranthus species could lead to seed bank dynamics during a five-year crop rotation. Weed rapid adaptation in new areas. Added to these characteristics Technol. 15: 170–176. Burnside, O. C., Wilson, R. G., Weisberg, S. and Hubbard, K. are the high growth rate and competitiveness, the ease of long G. 1996. Seed longevity of 41 weed species buried 17 years in distance pollen and seed dispersal, production of a large num- eastern and western Nebraska. Weed Sci. 44: 74–86. ber of viable seeds, the persistent seed bank, and an extended Coetzer, E., Al-Khatib, K. and Peterson, D. E. 2002. Glufosinate emergence pattern. efficacy on Amaranthus species in glufosinate-resistant soybean The expansion of A. tuberculatus var. rudis into new (Glycine max). Weed Technol. 16: 326–331. areas is expected to progress quickly after initial establish- Cole, H. E. and Holch, A. E. 1941. The root habits of certain ment and this should occur more rapidly than with monoe- weeds of southern Nebraska. Ecology 22: 141–147. cious amaranths. Given the level of the current infestation in Cordes, J. C. and Johnson, W. G. 2003. Common waterhemp southwestern Ontario, complete eradication is probably interference in corn. Proc. North Centr. Weed Sci. Soc. 43: 241 already difficult to achieve and re-infestation with new (Abstr.). plants from the United States is always possible. It might be Costea, M. 1996. The recording of some new adventive taxa for 41 possible to reduce or manage the impact of this invasive Romania in the Harbor of Constanta. Rev. Rom. Biol. Veg. : 91–96. weed if quick action is taken to identify and map the foci of Costea, M. 1998. Monograph of the genus Amaranthus L. in infestations of this weed, determine its population dynamics Romania. Ph.D. diss., University of Bucharest, College of Biology, under our climate and establish proper methods of control Bucharest, Romania. 210 pp. and management. Thorough cleaning of combines after har- Costea, M. and DeMason, D. A. 2001. Stem morphology and vest before moving equipment to other fields would limit anatomy in Amaranthus L. (Amaranthaceae): taxonomic signifi- the transport of seeds. Otherwise, it is probably only a mat- cance. J. Torrey Bot. Soc. 128: 254–281. ter of time until this species will become as problematic in Costea, M. and Tardif, F. J. 2003a. The name of the amaranth: Canada as it is now in the United States. histories of meaning. Sida 20: 1071–1081. Costea, M. and Tardif, F. J. 2003b. Conspectus and notes on the ACKNOWLEDGEMENTS genus Amaranthus (Amaranthaceae) in Canada. Rhodora 105: Suzanne Warwick and two anonymous reviewers provided 260–281. Costea, M. and Tardif, F. J. 2003c. The biology of Canadian suggestions that improved an earlier version of the manu- weeds 126. Amaranthus albus L., A. blitoides S. Watson and A. bli- script. Thanks are due to Rocio van der Laat and Micheal tum L. Can. J. Plant Sci. 83: 1039–1066. Owen for sharing their unpublished information about seed Costea, M. and Tardif, F. J. 2003d. The bracteoles in predation in common water-hemp. Alexandra Smith kindly Amaranthus (Amaranthaceae): their morphology, structure, func- assisted us with the scanning electron microscope. tion and taxonomic significance. Sida 20: 969–985. Costea, M., Sanders, A. and Waines, G. 2001a. Preliminary Aellen, P. 1959. Amaranthus L. Pages: 465–516 in G. Hegi, ed. results toward a revision of the species com- Illustrierte flora von Mitteleuropa, Vol. 3, Part. 2. Munich, Germany. plex (Amaranthaceae). Sida 19: 931–974. 520 CANADIAN JOURNAL OF PLANT SCIENCE

Costea, M., Waines, G. and Sanders, A. 2001b. Notes on some Hager, A. G. 1998. Waterhemp – a contemporary riddle for corn little known Amaranthus taxa (Amaranthaceae) in the United and soybean producers. Illinois Fertilizer Conference Proceedings. States. Sida 19: 975–992. 1998 Jan. 26–28. [Online] Available: http://frec.cropsci.uiuc.edu/ Costea, M., Waines, G. and Sanders, A. 2001c. Structure of the 1998/index.htm [2004 Aug.] pericarp in some Amaranthus L.(Amaranthaceae) species and its Hager, A. G., Wax, L. M., Simmons, F. W. and Stoller, E. W. taxonomic significance. Aliso 20: 51–60. 1997. Waterhemp management in agronomic crops: Univ. Ill. Bull. Costea, M., Weaver, S. E. and Tardif, F. J. 2004. Biology of No. 855. 12 pp. Canadian weeds 130. Amaranthus retroflexus L., A. powellii S. Hager, A. G., Wax, L. M., Bollero, G. A. and Simmons, F. W. Watson and A. hybridus L. Can. J. Plant Sci. 84: 631–668. 2002a. Common waterhemp (Amaranthus rudis Sauer) manage- Darbyshire, S. J. 2003. Inventory of Canadian agricultural weeds. ment with soil-applied herbicides in soybean (Glycine max (L.) Agriculture and Agri-Food Canada, Research Branch, Ottawa, ON. Merr.). Crop Prot. 21: 277–283. 396 pp. [Online] Available: http://res2.agr.ca/ecorc/weeds_herbes Hager, A. G., Wax, L. M., Stoller, E. W. and Bollero, G. A. /title-titre_e.htm [August 2004]. 2002b. Common waterhemp (Amaranthus rudis) interference in Douglas, G. W., Straley, G. B., Meidinger, D. and Pojar. J. soybean. Weed Sci. 50: 607–610. 1998. Illustrated flora of British Columbia. Vol. 1. Gymnosperms Hager, A. G., Wax, L. M., Bollero, G. A. and Stoller, E. W. and Dicotyledons (Aceraceae through Asteraceae). British 2003. Influence of diphenylether herbicide application rate and Columbia Ministry of Environment, Lands and Parks and Ministry timing on common waterhemp (Amaranthus rudis) control in soy- of Forests, Victoria, BC. 436 pp. bean (Glycine max). Weed Technol. 17: 14–20. Dirks, J. T., Johnson, W. G., Smeda, R. J., Wiebold, W. J. and Hartzler, R. G., Buhler, D. D. and Stoltenberg, D. E. 1999. Massey, R. E. 2000a. Reduced rates of sulfentrazone plus chlo- Emergence characteristics of four annual weed species. Weed Sci. rimuron and glyphosate in no-till, narrow-row, glyphosate-resis- 47: 578–584. tant Glycine max. Weed Sci. 48: 618–627. Hartzler, R. G., Battles, B. A. and Nordby, D. 2004. Effect of Dirks, J. T., Johnson, W. G., Smeda, R. J., Wiebold, W. J. and common waterhemp (Amaranthus rudis) emergence date on Massey, R. E. 2000b. Use of pre-plant sulfentrazone in no-till, nar- growth and fecundity in soybean. Weed Sci. 52: 242–245. row-row, glyphosate-resistant Glycine max. Weed Sci. 48: 628–639. Heap, I. 2004. International survey of herbicide resistant weeds. Evans, S. P., Knezevic, S. Z., Lindquist, J. L. and Shapiro, C. [Online] Available: http://www.weedscience.com [2004 Aug.]. A. 2003. Influence of nitrogen and duration of weed interference Hellwig, K. B., Johnson, W. G. and Massey, R. E. 2003. Weed on corn growth and development. Weed Sci. 51: 546–556. management and economic returns in no-tillage herbicide-resistant Falk, J. S., Shoup, D. E., Al-Khatib, K. and Peterson, D. E. corn (Zea mays). Weed Technol. 17: 239–248. 2004. Survey of common waterhemp resistance to protox- and Hinz, J. R. R. and Owen, M. D. K. 1997. Acetolactate synthase ALS-inhibiting herbicides in northeast Kansas. Weed Science resistance in a common waterhemp (Amaranthus rudis) popula- Society of America Annual Meeting, 2004 Feb. 07–11. Kansas tion. Weed Technol. 11: 13–18. City, MO. Vol. 44, p. 51 (Abstr.). Holmgren, P. K., Holmgren, N. H. and Barnett, L. C. 1990. Felix, J. and Owen, M. D. K. 1999. Weed population dynamics in Index Herbariorum. Part I: The herbaria of the world. 8th ed. New land removed from conservation reserve program. Weed Sci. 47: York Botanical Garden, Bronx, NY. 693 pp. 511–517. Horak, M. J. and Loughin, T. M. 2000. Growth analysis of four Foes, M. J., Liu, L., Tranel, P. J., Wax, L. M. and Stoller, E. W. Amaranthus species. Weed Sci. 48: 347–355. 1998. A biotype of common waterhemp (Amaranthus rudis) resis- Horak, M. J. and Peterson, D. E. 1995. Biotypes of Palmer ama- tant to triazine and ALS herbicides. Weed Sci. 46: 514–520. ranth (Amaranthus palmeri) and common waterhemp (Amaranthus Franssen, A. S., Skinner, D. Z., Al-Khatib K., Horak, M. J. and rudis) are resistant to imazethapyr and thifensulfuron. Weed Kulakow, P. A. 2001a. Interspecific hybridization and gene flow Technol. 9: 192–195. of ALS resistance in Amaranthus species. Weed Sci. 49: 598–606. Hoss, N. E., Al-Khatib, K., Peterson, D. E. and Loughin, T. M. Franssen, A. S., Skinner, D. Z., Al-Khatib, K. and Horak, M. J. 2003. Efficacy of glyphosate, glufosinate, and imazethapyr on 2001b. Pollen morphological differences in Amaranthus species selected weed species. Weed Sci. 51: 110–117. and interspecific hybrids. Weed Sci. 49: 732–737. Invaders Database System. 2003. Agricultural Research Service Galatowitsch, S. M. and van der. Valk, A. G. 1996. The vegetation U.S.D.A. [Online] Available: http://invader.dbs.umt.edu. [2004 Aug.]. of restored and natural prairie wetlands. Ecol. Appl. 6: 102–112. Jeschke, M. R., Tranel, P. J. and Rayburn, A. L. 2003. DNA Grant, W. F. 1959. Cytogenetic studies in Amaranthus. I. content analysis of smooth pigweed (Amaranthus hybridus) and Cytological aspects of sex determination in dioecious species. Can. tall waterhemp (A. tuberculatus): implications for hybrid detection. J. Bot. 37: 413–417. Weed Sci. 51: 1–3. Greene, H. C. 1945. Notes on Wisconsin parasitic fungi. VII. Am. Joel, D. M. and Liston, A. 1986. New adventive weeds in Israel. Midl. Naturalist 34: 258–270. Israel J. Bot. 35: 215–223. Gregory, P. H. 1973. The microbiology of the atmosphere. 2nd Johnson, W. G., Bradley, P. R., Hart, S. E., Buesinger, M. L. ed. John Wiley & Sons, New York, NY. 377 pp. and Massey, R. E. 2000. Efficacy and economics of weed man- Greuter, W., McNeill, J., Barrie, F. R., Burdet, H. M., Demoulin, agement in glyphosate-resistant corn (Zea mays). Weed Technol. V., Filgueiras, T. S., Nicolson, D. H., Silva, P. C., Skog, J. E., 14: 57–65. Trehane, P., Turland, N. J. and Hawksworth, D. L. 2001. Jones, S. M., Smeda, R. J., Smith, G. S. and Johnson, W. G. 1998. International code of botanical nomenclature (St Louis Code). The effect of waterhemp competition on soybean yield: Proc. North [Online] Available:http://www.bgbm.org/IAPT/Nomenclature/ Centr. Weed Sci. Soc. 53: 146 (Abst.). Code/SaintLouis/0000St.Luistitle.htm [2004 Aug.] Jurik, T. W. and Zhang ShuYu. 1999. Tractor wheel traffic effects Guo, P. and Al-Khatib, K. 2003. Temperature effects on germi- on weed emergence in central Iowa. Weed Technol. 13: 741–746. nation and growth of redroot pigweed (Amaranthus retroflexus), Krausz, R. F., Kapusta, G. and Mathews, J. L. 1998. Sulfentrazone Palmer amaranth (A. palmeri), and common waterhemp (A. rudis). for weed control in soybean (Glycine max). Weed Technol. 12: Weed Sci. 51: 869–875. 684–689. COSTEA ET AL. — AMARANTHUS TUBERCULATUS 521

Krausz, R. F. and Young, B. G. 2003. Sulfentrazone enhances Peterson, D. E. 1999. The impact of herbicide-resistant weeds on weed control of glyphosate in glyphosate-resistant soybean Kansas agriculture. Weed Technol. 13: 632–635. (Glycine max). Weed Technol. 17: 249–255. Pfeifer, T. L., Cho, I., Gibson, D. J., Young, B. G. and Wood, Kyaw Tha Paw, U. and Hotton, C. 1989. Optimum pollen and A. J. 2001. Utility of trigonelline as a biochemical marker for female receptor size for anemophily. Am. J. Bot. 76: 445–453. interspecific competition between soybean and weed common Lemen, C. 1980. Allocation of reproductive effect to the male and waterhemp. Biol. Plant. 44: 619–622. female strategies in wind-pollinated plants. Oecologia 45: 156–159. Pratt, D. B. and Clark, L. G. 2001. Amaranthus rudis and A. tuber- Leon, R. G. and Owen, M. D. K. 2003. Regulation of weed seed culatus—one species or two? J. Torrey Bot. Soc. 128: 282–296. dormancy through light and temperature interactions. Weed Sci. Preston, D. A. 1945. Host index of Oklahoma plant diseases. 51: 752–758. Oklahoma Agric. Coll. Agric. Exp. Sta. Techn. Bull. T 21: 1–168. Leon, R. G., Knapp, A. D. and Owen, M. D. K. 2004. Effect of Preston, D. A. and Dostall, L. 1955. Minnesota Plant Diseases. temperature on the germination of common waterhemp USDA, Plant Diseases Epidemiology and Identification Section, (Amaranthus tuberculatus), giant foxtail (Setaria faberi), and vel- Special Publication 8: 1–184. vetleaf (Abutilon theophrasti). Weed Sci. 52: 67–73. Quinn, J. A., Bram, M. R. and Taylor, T. E. 2000. Female Lewis, H. W., Vinay, P. and Zenger, E. V. 1983. Airborne and resource allocation in response to pollen availability in plants from allergenic pollen of North America. The Johns Hopkins University freshwater and salt marsh populations of Amaranthus cannabinus. Press. Baltimore and London, 254 pp. J. Torrey Bot. Soc. 127: 83–86. Li, J., Smeda, R. J., and Wait, J. D. 2004a. Population differ- Riddell, J. L. 1835. Synopsis of the flora of the western states. W. ences in waterhemp treated with glyphosate and lactofen. Weed J. Med. Phys. Sci. 41: 367. Science Society of America Annual Meeting. 2004 Feb. 07–11. Roehl, S. R., Bredehoeft, M. and Fischer, J. 2001. Efficacy and Kansas City, MO. Vol. 44, p. 15 (Abstr.). economic viability of current and potential weed control options at Li, J., Smeda, R. J., Nelson, K. A. and Dayan, F. E. 2004b. Southern Minnesota Beet Sugar Cooperative. Proc. 31st Biennial Physiological basis for resistance to diphenyl ether herbicides in Meeting of the Am. Soc. Sugar Beet Technologists, pp. 72–74. common waterhemp (Amaranthus rudis). Weed Sci. 52: 333–338. Sauer, J. D. 1955. Revision of the dioecious amaranths. Madroño Lovell, S. T., Wax, L. M., Horak, M. J. and Peterson, D. E. 13: 5–46. 1996. Imidazolinone and sulfonylurea resistance in a biotype of Sauer, J. D. 1972. The dioecious amaranths: a new species name common waterhemp (Amaranthus rudis). Weed Sci. 44: 789–794. and major range extensions. Madroño 21: 426–434. Mayo, C. M., Horak, M. J., Peterson, D. E. and Boyer, J. E. 1995. Sellers, B. A., Smeda, R. J., Johnson, W. G., Kendig, J. A. and Differential control of four Amaranthus species by six postemergence Ellersieck, M. R. 2003. Comparative growth of six Amaranthus herbicides in soybean (Glycine max). Weed Technol. 9: 141–147. species in Missouri. Weed Sci. 51: 329–333. Menalled, F. D., Liebman, M. and Buhler, D. 2004. Impact of com- Shoup, D. E., Al-Khatib, K. and Peterson, D. E. 2003. Common posted swine manure and tillage on common waterhemp (Amaranthus waterhemp (Amaranthus rudis) resistance to protoporphyrinogen rudis) competition with soybean. Weed Sci. 52: 605–613. oxidase-inhibiting herbicides. Weed Sci. 51: 145–150. Moquin-Tandon, C. H. B. A. 1849. The genus Acnida. Pages Smith, D. A. and Hallett, S. G. 2004a. Epidemiology of the can- 277–278 in A. P. DeCandolle, ed. Prodromus systematis naturalis didate bioherbicide Microsphaeropsis amaranthi infecting water- regni vegetabilis 13(2), Paris, France. hemp, Amaranthus tuberculatus. Weed Science Society of Mosyakin, L. S. and Robertson, R. K. 1996. New infrageneric America Annual Meeting. 2004 Feb. 07–11. Kansas City, MO. taxa and combinations in Amaranthus (Amaranthaceae). Ann. Bot. Vol. 44, p. 74. (Abstr.). Fenn. 33: 275–281. Smith, D. A. and Hallett, S. G. 2004b. Interactions between Murray, M. J. 1940. The genetics of sex determination in the fam- Microsphaeropsis amaranth and glyphosate for the control of ily Amaranthaceae. Genetics 25: 409–431. waterhemp, Amaranthus tuberculatus. Weed Science Society of Niekamp, J. W. and Johnson, W. G. 2001. Weed management America Annual Meeting. 2004 Feb. 07–11. Kansas City, MO. with sulfentrazone and flumioxazin in no-tillage soyabean Vol. 44, p. 73. (Abstr.). (Glycine max). Crop Prot. 20: 215–220. Sprague, C. L., Stoller, E. W. and Wax, L. M. 1997a. Response Nolte, S. A. and Young, B. G. 2002a. Efficacy and economic of an acetolactate synthase (ALS)–resistant biotype of Amaranthus return on investment for conventional and herbicide-resistant corn rudis to selected ALS-inhibiting and alternative herbicides: Weed (Zea mays). Weed Technol. 16: 371–378. Res. 37: 93–101. Nolte, S. A. and Young, B. G. 2002b. Efficacy and economic Sprague, C. L., Stoller, E. W., Wax, L. M. and Horak, M. J. return on investment for conventional and herbicide-resistant soy- 1997b. Palmer amaranth (Amaranthus palmeri) and common bean (Glycine max). Weed Technol. 16: 388–395. waterhemp (Amaranthus rudis) resistance to selected ALS-inhibit- Nordby, D. E. and Hartzler, R. G. 2004. Influence of corn on ing herbicides. Weed Sci. 45: 192–197. common waterhemp (Amaranthus rudis) growth and fecundity. Steckel, L. E. and Sprague, C. L. 2004. Common waterhemp Weed Sci. 52: 255–259. (Amaranthus rudis) interference in corn. Weed Sci. 52: 359–364. Patzoldt, W. L., Tranel, P. J. and Hager, A. G. 2002. Variable Steckel, L. E., Sprague, C. L. and Hager, A. G. 2002. Common herbicide responses among Illinois waterhemp (Amaranthus rudis waterhemp (Amaranthus rudis) control in corn (Zea mays) with and A. tuberculatus) populations. Crop Prot. 21: 707–712. single preemergence and sequential applications of residual herbi- Patzoldt, W. L., Dixon, B. S. and Tranel, P. J. 2003. Triazine cides. Weed Technol. 16: 755–761. resistance in Amaranthus tuberculatus (Moq.) Sauer that is not Steckel, L. E., Sprague, C. L., Stoller, E. W. and Wax, L. M. site-of-action mediated. Pest Management Sci. 59: 1134–1142. 2004. Temperature effects on germination of nine Amaranthus Patzoldt, W. L., Hager, A. G. and Tranel, P. J. 2004. Evaluation species. Weed Sci. 52: 217–221. of glyphosate responses in tall waterhemp (Amaranthus tubercula- Steckel, L. E., Sprague, C. L., Hager, A. G., Simmons, F. W. tus) using a quantitative genetics approach. Weed Science Society and Bollero, G. A. 2003. Effects of shading on common water- of America Annual Meeting. 2004 Feb. 07–11. Kansas City, MO. hemp (Amaranthus rudis) growth and development. Weed Sci. 51: Vol. 44, p. 66 (Abstr.). 898–903. 522 CANADIAN JOURNAL OF PLANT SCIENCE

Steinau, A. N., Skinner, D. Z. and Steinau, M. 2003. Mechanism Uline, E. B. and Bray, W. L. 1895. Synopsis of North American of extreme genetic recombination in weedy Amaranthus hybrids. Amaranthaceae. II. Bot. Gaz. 20: 155–161. Weed Sci. 51: 696–701. Umbanhowar, C. E., Jr. 1992. Abundance, vegetation, and envi- Stevens, O. A. 1932. The number and weight of seeds produced by ronment of four patch types in a northern mixed prairie. Can. J. the weeds. Am. J. Bot. 19: 784–794. Bot. 70: 277–284. Sweat, J. K., Horak, M. J., Peterson, D. E., Lloyd, R. W. and USDA, NRCS. 2004. The PLANTS Database. Version 3.5. Boyer, J. E. 1998. Herbicide efficacy on four Amaranthus species National Plant Data Center, Baton Rouge, LA. [Online] Available: in soybean (Glycine max). Weed Technol. 12: 315–321. http://plants.usda.gov (2004 Aug.). Thellung, A. 1914. Amaranthus L. Pages 225–356 in P. Ascherson Wetzel, D. K., Horak, M. J. and Skinner, D. Z. 1999a. Use of and P. Graebner, eds. Synopsis der Mitteleuropaischen flora. Vol. PCR-based molecular markers to identify weedy Amaranthus 5. Leipzig. species. Weed Sci. 47: 518–523. Tranel, P. J. and Wright, T. R. 2002. Resistance of weeds to Wetzel, D. K., Horak, M. J., Skinner, D. Z. and Kulakow, P. A. ALS-inhibiting herbicides: what have we learned? Weed Sci. 50: 1999b. Transfer of herbicide resistance traits from Amaranthus 700–712. palmeri to A. rudis. Weed Sci. 47: 538–543. Tranel, P. J., Wassom, J. J., Jeschke, M. R. and Rayburn, A. L. Wilson W. G. 1908. Studies in North American Peronosporales— 2002. Transmission of herbicide resistance from a monoecious to IV. Host index. Bull. Torrey Bot. Club. 11: 543–554. a dioecious weedy Amaranthus species. Theor. Appl. Genet. 105: Weed Science Society of America. 2004. Common/Latin weed 674–679. names. [Online] Available: http://www.wssa.net/weednames/com- Tranel, P. J., Trucco, F., Tatum, T. C. and Rayburn, L. A. 2004. mon.asp [2004 Aug.]. Molecular, cytogenetic and morphological analysis of smooth pig- Young, B. G., Young, J. M., Gonzini, L. C., Hart, S. E., Wax, weed (Amaranthus hybridus) × waterhemp (Amaranthus tubercula- L. M. and Kapusta, G. 2001. Weed management in narrow- and tus) BC1 progeny. Weed Science Society of America Annual Meeting. wide-row glyphosate-resistant soybean (Glycine max). Weed 2004 Feb. 07–11. Kansas City, MO. Vol. 44, p. 229. (Abstr.) Technol. 15: 112–121. Trucco, F., Tranel, P. J., Jeschke, M. R. and Rayburn, L. A. 2004. Field hybridization frequencies between monoecious smooth pigweed (Amaranthus hybridus) and dioecious waterhemp (Amaranthus tuber- culatus). Weed Science Society of America Annual Meeting. 2004 Feb. 07–11. Kansas City, MO. Vol. 44, p. 229. (Abstr.).